Comment 415 for bug 532633

Mark:

Good, finally some guidance. You shouldn't wait for people to think up data products on their own, that risks people spinning their wheels creating "data" that gets discarded because it doesn't meet your definition and leads to people feeling they are being ignored.

You and the design team are the only group in a position define what is acceptable data in your decision-making process. Putting forward some questions you want answered like you did above is helpful. But you could go further, and articulate a framework by which questions can be proposed by externals, accepted by the design team as important to the design process, and then answered with an acceptable data collection methodology.

If you do not articulate a data feedback framework that is acceptable to the design team then how is anyone suppose to know what you think is and is not acceptable? If you don't have a process by which people can propose questions worth answering with data, how do people know what to collect data on?

Generally speaking..when doing an experiment in a professional research setting you have to have a firm grasp on the questions you want answered and how you plan to collect data _before_ you do the experiment. You seldom just throw stuff together and "see what happens." Neither of the questions nor the data collection methodology were communicated before this "experiment" with the button positions. Something to think about before you embark on the next round of design experiments.